The Design is aimed toward retaining the dexterity of a human while remaining simple and cheap; to accomplish this we have a linear actuator that allows a human scale 7-DOF arm to move from 4 inches off the ground to 3 1/2 feet up. Mobility is made simple by using two wheels, ball casters ( for stability ), and a 1 1/2 foot wide base. This allows the 4 foot tall robot to maneuver across tile and carpet, but more importantly- it fits through a door with ease. In addition, the flat face allow the machine to carry objects through said doors and turn without serious issues. The sectioned spaces within the machine are ordered as misc, motors, electronics, and batteries from top to bottom. The intention is to allow the heavy pulling motors actuate screw gear sets to reduce their speed and lock into position to prevent unwanted motion. 4 motors within the main body can then transfer motion via bicycle like cables to the arm ( 2 at the shoulder, 2 at the wrist ) leaving 3 motors in the arm itself ( 2 for rotation, 1 for gripping ) making for a safe and light arm. The face of the robot ( the top rounded corner of the flat face, I did not place this onto the model ) contains a set of rotatable stereoscopic cameras, these are the most complex sensors on the machine, I will explain this in the next step. Step 2: The Customer and Machine How does this machine work to provide me with a service? It all begins with an operator, this person carries the complicated and expensive equipment and makes the complicated decisions. By scanning the intended 'work space' and declaring object structure and object function- an operator can categorize the robot's world. By using programming similar to SLAM (uses previously measured objects to find them in space surprisingly accurately and reliably) and categorizing objects into categories such as permanent (walls, ceiling fans), generally static (furniture), temporary (cups, trash, etc), door handles, machine interfaces, etc. we can simplify one of the complicated jobs of today's machines- understanding where it is in relation to other objects. The operator can now dump this information into a robot and leave the location. Now the robot simply takes orders and breaks them into simple actions. For example: clean the cups, to accomplish task the machine must locate cups using only stereoscopic vision and the information it has and has gathered ( cups are often found in the bedroom and kitchen ) and sends a move command to move toward one. Then a different program can be activated to pick up the cup, followed by a move command to return it to the kitchen sink. All tasks can be broken down into simpler tasks that have been preprogrammed or machined learned before distribution. All of this together allows us to lower the price of the machine and the time it takes to process its tasks while improving the reliability of the machine so that it does not break all of your china while your not home. The Customer only sees the operator once, from there he or she only has to give their machine a set of tasks such as get the papers from room () or clean the dishes while I am gone. Preferably this machine would operate when there is little activity in its work space as to reduce frustration and offer opportunity where nobody would be active to complete tasks such as cleaning, organizing, preparing, and even protecting a house while the family is gone by imitating activity within the house. Now many people can afford a novel yet useful machine within their house today without the need for technology that will only come in many more years in order to make house machines more interactive, capable, useful, and in the customer's eye, necessary ( just as our modern appliances ).
Participated in the
Design for Robotics Contest 2017